Said Safi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Said Safi has a total of 135 co-authors. Here are the top co-authors by number of shared publications:
- Miloud Frikel
ENSICAEN, Caen, France
11 shared publications - Rachid Fateh
Université Ibn Zohr
11 shared publications - Anouar Darif
Sultan Moulay Slimane University
8 shared publications - Miloud Frikel
Greyc UMR 6072 CNRS ENSICAEN Caen France
7 shared publications Ahmed Boumezzough
Sultan Moulay Slimane University
5 shared publications- Mathieu Pouliquen
Unicaen, Caen, France
5 shared publications
Latest Publications
- 10 November 2026
- Published by Springer Nature in Circuits, Systems, and Signal Processing
- p. 1-22
Adaptive Frequency Selection for Wideband DOA Estimation: An Enhanced TOFS Approach
- 10 November 2026
- Published by Springer Nature in Circuits, Systems, and Signal Processing
- p. 1-22
Adaptive Frequency Selection for Wideband DOA Estimation: An Enhanced TOFS Approach
- Open Access
- 31 July 2026
- Published by Elsevier in Information Processing & Management
- Vol. 64 (1)
Bridging the skills gap in recruitment: A RAG-based LLM framework for cybersecurity job advertisement analysis
- 31 July 2026
- Published by Elsevier in Information Processing & Management
- Vol. 64 (1)
Open AccessBridging the skills gap in recruitment: A RAG-based LLM framework for cybersecurity job advertisement analysis
- 1 April 2026
- Published by Springer Nature
- p. 413-417
Revealing CDN-Side Cybersecurity Risks in HTTPS Connections
- 1 April 2026
- Published by Springer Nature
- p. 413-417
Revealing CDN-Side Cybersecurity Risks in HTTPS Connections
- 15 March 2026
- Published by Taylor & Francis in EDPACS
- p. 1-8
Artificial intelligence in cybersecurity: Addressing ethical considerations and legal implications
1. In today’s digital era, protecting information and ensuring online security have become critical. Cybersecurity involves implementing strategies and practices to defend computer systems and netw...
- 15 March 2026
- Published by Taylor & Francis in EDPACS
- p. 1-8
Artificial intelligence in cybersecurity: Addressing ethical considerations and legal implications
1. In today’s digital era, protecting information and ensuring online security have become critical. Cybersecurity involves implementing strategies and practices to defend computer systems and netw...
- 22 February 2026
- Published by Springer Nature in Telecommunication Systems
- Vol. 89 (1) , 40
Kernel-Based Machine Learning Ellipsoidal Outer Bounding for Non-Line-of-Sight Outdoor and Indoor Channel Identification
- 22 February 2026
- Published by Springer Nature in Telecommunication Systems
- Vol. 89 (1) , 40
Kernel-Based Machine Learning Ellipsoidal Outer Bounding for Non-Line-of-Sight Outdoor and Indoor Channel Identification
- Open Access
- 30 October 2025
- Published by Springer Nature in Research Square
A Hybrid Framework for Blind Channel Estimation and Equalization in Multipath OFDMA Systems
This paper addresses the problem of blind channel estimation and equalization in orthogonal frequency-division multiple access (OFDMA) systems over multipath Rayleigh fading channels, modeled according to BRAN-A, BRAN-B, and Proakis B profiles. We investigate and compare three algorithms: the classical Fourth-Order Cumulant method based on higher-order statistics (HOS), a deep learning approach using an autoencoder, and a novel hybrid technique combining cumulant-derived statistical features with autoencoder learning. The Fourth-Order Cumulant method leverages higher-order statistical moments to extract channel information without requiring pilot signals, making it computationally lightweight and analytically interpretable. It performs well in stationary or low-noise conditions with slow channel variations. However, its performance degrades in the presence of high noise, nonlinearities, or rapid fading due to its reliance on rigid statistical assumptions. The autoencoder-based approach employs a neural network to implicitly learn channel effects directly from the magnitude or complex samples of the received signal, without explicit channel modeling. This method offers high adaptability and robustness to nonlinear distortions and time-varying fading, significantly outperforming cumulant-based estimation in complex scenarios. However, it incurs higher computational costs and reduced transparency, as its learned parameters are not directly interpretable. The proposed hybrid method integrates cumulant-based features into the autoencoder’s input space, enabling the network to benefit from both the explicit statistical structure provided by HOS and the data-driven learning capacity of deep neural models. Simulations conducted over a wide range of signal-to-noise ratio (SNR) values and under realistic multipath fading conditions demonstrate that: The cumulant method is optimal for low-noise, slowly varying channels, The autoencoder provides robust performance in nonlinear or rapidly changing environments, and, The hybrid approach consistently achieves the lowest mean square error (MSE) and bit error rate (BER), combining the interpretability and efficiency of cumulants with the adaptability of deep learning. In conclusion, this work highlights the complementarity between statistical and deep learning methods for blind channel estimation and equalization, demonstrating that hybridization can yield a balanced and powerful solution for next-generation wireless systems, such as OFDMA.
- 30 October 2025
- Published by Springer Nature in Research Square
Open AccessA Hybrid Framework for Blind Channel Estimation and Equalization in Multipath OFDMA Systems
This paper addresses the problem of blind channel estimation and equalization in orthogonal frequency-division multiple access (OFDMA) systems over multipath Rayleigh fading channels, modeled according to BRAN-A, BRAN-B, and Proakis B profiles. We investigate and compare three algorithms: the classical Fourth-Order Cumulant method based on higher-order statistics (HOS), a deep learning approach using an autoencoder, and a novel hybrid technique combining cumulant-derived statistical features with autoencoder learning. The Fourth-Order Cumulant method leverages higher-order statistical moments to extract channel information without requiring pilot signals, making it computationally lightweight and analytically interpretable. It performs well in stationary or low-noise conditions with slow channel variations. However, its performance degrades in the presence of high noise, nonlinearities, or rapid fading due to its reliance on rigid statistical assumptions. The autoencoder-based approach employs a neural network to implicitly learn channel effects directly from the magnitude or complex samples of the received signal, without explicit channel modeling. This method offers high adaptability and robustness to nonlinear distortions and time-varying fading, significantly outperforming cumulant-based estimation in complex scenarios. However, it incurs higher computational costs and reduced transparency, as its learned parameters are not directly interpretable. The proposed hybrid method integrates cumulant-based features into the autoencoder’s input space, enabling the network to benefit from both the explicit statistical structure provided by HOS and the data-driven learning capacity of deep neural models. Simulations conducted over a wide range of signal-to-noise ratio (SNR) values and under realistic multipath fading conditions demonstrate that: The cumulant method is optimal for low-noise, slowly varying channels, The autoencoder provides robust performance in nonlinear or rapidly changing environments, and, The hybrid approach consistently achieves the lowest mean square error (MSE) and bit error rate (BER), combining the interpretability and efficiency of cumulants with the adaptability of deep learning. In conclusion, this work highlights the complementarity between statistical and deep learning methods for blind channel estimation and equalization, demonstrating that hybridization can yield a balanced and powerful solution for next-generation wireless systems, such as OFDMA.
- 2 July 2025
- Published by Springer Nature
- p. 16-29
Kernel Adaptive Filters for Machine Learning-Based System Identification with Binary Output Data
- 2 July 2025
- Published by Springer Nature
- p. 16-29
Kernel Adaptive Filters for Machine Learning-Based System Identification with Binary Output Data
- Open Access
- 1 April 2025
- Published by Institute of Advanced Engineering and Science in Indonesian Journal of Electrical Engineering and Computer Science
- Vol. 38 (1) , 215-226
Modeling non-linear communication systems using neural networks
Nonlinear systems present significant modeling challenges due to their complex dynamics and often unpredictable behavior. Traditional mathematical approaches can struggle to represent such systems accurately. In recent years, neural networks have emerged as promising tools to address this challenge. This article explores the use of neural networks to model nonlinear systems, focusing specifically on the application of the Hammerstein system. We examine network architecture and training methodologies suited to the complexity of nonlinear dynamics. Additionally, we explore strategies to improve the interpretability of neural network models in this context, enabling a better understanding of the underlying behavior of the system. Through a case study and empirical evaluations, we demonstrate the effectiveness of neural network-based approaches for estimating the behavior of nonlinear systems. Our work highlights the potential of neural networks as a versatile and powerful tool for modeling complex nonlinear phenomena.
- 1 April 2025
- Published by Institute of Advanced Engineering and Science in Indonesian Journal of Electrical Engineering and Computer Science
- Vol. 38 (1) , 215-226
Open AccessModeling non-linear communication systems using neural networks
Nonlinear systems present significant modeling challenges due to their complex dynamics and often unpredictable behavior. Traditional mathematical approaches can struggle to represent such systems accurately. In recent years, neural networks have emerged as promising tools to address this challenge. This article explores the use of neural networks to model nonlinear systems, focusing specifically on the application of the Hammerstein system. We examine network architecture and training methodologies suited to the complexity of nonlinear dynamics. Additionally, we explore strategies to improve the interpretability of neural network models in this context, enabling a better understanding of the underlying behavior of the system. Through a case study and empirical evaluations, we demonstrate the effectiveness of neural network-based approaches for estimating the behavior of nonlinear systems. Our work highlights the potential of neural networks as a versatile and powerful tool for modeling complex nonlinear phenomena.
- 31 May 2024
- Published by Springer Nature in International Journal of Speech Technology
- Vol. 27 (2) , 329-339
An experiment of Moroccan dialect speech recognition in noisy environments using PocketSphinx
- 31 May 2024
- Published by Springer Nature in International Journal of Speech Technology
- Vol. 27 (2) , 329-339
An experiment of Moroccan dialect speech recognition in noisy environments using PocketSphinx
- 13 February 2024
- Published by Springer Nature in Cluster Computing
- Vol. 27 (5) , 5867-5877
5 G and beyond wireless network optimization through RIS technology: a pricing game
- 13 February 2024
- Published by Springer Nature in Cluster Computing
- Vol. 27 (5) , 5867-5877
5 G and beyond wireless network optimization through RIS technology: a pricing game